WiFi Password Safety

Question

Are there any best practices regarding password safety? The following is taken from a wifi tutorial sketch.

#include <SPI.h>
#include <WiFi.h>

char ssid[] = "yourNetwork";      // your network SSID (name) 
char pass[] = "secretPassword";   // your network password
int keyIndex = 0;                 // your network key Index number (needed only for WEP)

While I accept that the tutorial is there for learning how to use the shield, I would assume hard coding the password into the sketch leaves this open to attackers. Is this actually the case? If so, are there any strategies to harden the sketch, for example with encrypted storage?

I am a relative newbie with physical computing and Arduino, so sorry if I am overlooking something obvious.

Answer 1

You're actually entering into a very tricky and difficult area.

Yes, having a plain text password in your sketch is a risk. It is stored in flash as plain text. If an attacker can gain access to the contents of the flash through some means (whatever they may be) then they could obtain that WiFi password.

You would think that encrypting the password would be the answer, but that's not the case. The problem with an encrypted password like this is it has to be decrypted to use it. That means that both the encrypted password and the encryption key are both stored together in flash. If you can get one you can get the other, and thus the encryption is null and void. Encryption security relies on keeping the encrypted data and the key used to encrypt/decrypt it completely separate. After all, you don't leave the key to your house in the front door lock when you go out, do you? It's exactly the same.

So what do you do?

Well, there's lots of different ways of looking at security. In this case there is one specific viewpoint that works best:

Security Through Obscurity

Your system is (pretty much) unique. Nobody but you knows just what it does and how it does it, or how you interact with it. It's not obvious to the casual observer just what it is or what could be done with it. That means that they won't know how to break in to it.

The majority of use of vulnerabilities in systems rely on those systems being ubiquitous - that is, all over the place; easy to obtain. In short, easy to find out what they are, how they work, and even reverse engineer them to a certain extent. Attackers aren't going to be able to do that with your system, since you are the only person in the world with it.

However, it's far from perfect, and you shouldn't rely only on Security Through Obscurity.

Simple obfuscation of the password (not actual encryption, but just storing it in a way that isn't obvious) can help, since it isn't then easy to see where in a memory dump the password would be. Splitting it up and scattering it around the place, too, can be useful.

Basically there is no simple answer, other than make it as hard for an attacker to find sensitive things as possible and don't tell anyone what you have done to hide it ;)

Answer 2

Unfortunately, I don't have a precise answer to your question currently, but I would like to warn you about the Security Through Obscurity approach proposed by Majenko. This approach may be reasonable for something very homemade, but it is strongly discouraged by researchers and experts.

For instance, if you have a look at the Guide to General Server Security you will read between the "Server Security Principle":

Open Design—System: security should not depend on the secrecy of the implementation or its components.

So, try always to use a standard approach if you can.